lkml.org 
[lkml]   [2002]   [Sep]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
Subjectirman takes 50x longer without O(1) on uniprocessor
From
I've noticed running irman on uniprocessor with the 
O(1) scheduler takes a lot less time than the mainline
scheduler. Only 2.4.20-pre5 and 2.4.20-pre6 below don't
have a version of the O(1) scheduler:

seconds to run irman 3 times
2.4.20-pre4-ac1 1420
2.4.20-pre5-ac1 1872
2.4.20-pre5 162088
2.4.20-pre5aa1 2264
2.4.20-pre5aa2 2274
2.4.20-pre6 111651
2.5.32-viro-mm1 2209
2.5.33-mm1-poll 2168
2.5.33-mm1 1679
2.5.33-mm5 2374
2.5.33 2408

There are some differences in context switch, user, and
system times between O(1) haves and have nots. Oddly, the
user time without O(1) is higher, yet it takes longer to
complete.

vmstat 60 on 2.4.20-pre6
procs memory swap io system cpu
r b w swpd free buff cache si so bi bo in cs us sy id
3 0 0 2012 364212 10468 2264 0 0 0 1 100 27467 18 82 0
2 0 0 2012 364172 10492 2268 0 0 0 1 100 27488 18 82 0
2 0 0 2012 364144 10508 2268 0 0 0 1 100 27555 18 82 0

vmstat 60 on 2.4.20-pre5-ac1
procs memory swap io system cpu
r b w swpd free buff cache si so bi bo in cs us sy id
3 0 0 1692 365848 1488 7756 0 0 0 1 100 36821 4 96 0
2 0 0 1692 365832 1504 7756 0 0 0 1 100 37334 6 94 0
3 0 0 1692 365816 1520 7756 0 0 0 1 100 37005 5 95 0

Quad xeon doesn't have a huge difference in "real"
time to run irman.

Does anyone know what would cause a 50-100x difference in time
to execute irman on uniprocessor?

--
Randy Hron
http://home.earthlink.net/~rwhron/kernel/bigbox.html

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:28    [W:0.034 / U:1.652 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site